When last we met our hero, he was trying various things to get people aware and ready to buy his novel Entropy. Among those was an idea to create a couple Twitter accounts for the two main characters from the novel. I figured they could build a following among like-minded “people” in the shadowy world of BDSM Twitter. And those people might be interested in reading the book. Turns out: not so much.
Building a following is relatively easy. I built up the two timelines using a mix of stuff from the book, and original material. So if you went there, there was something interesting to read. About half the people I follow from those accounts follow back. So I got them both up to about 300 followers in a few days. A couple interesting things about this part of the exercise: I followed exactly the same people, both men and women, from each account. Sir ended up with 325 followers, but Lisa only got 292. There was an equal amount of good content on both accounts. This is actually pretty puzzling to me (conventional wisdom is that chicks get follow-backs more readily than dudes), and I wonder if I would have seen the same result if they had actual humans in the AVIs instead of art.
The other interesting thing was the DMs I received. Both accounts got direct messages from people, but Lisa got a lot more. There were both men trying to start some kind of a relationship, and women offering to give love and support. I’m tweeting in character on the accounts, so I thought I kind of should DM back in character. Except that’s just too weird and creepy. So instead, I simply replied to the DMs with the tweet I have pinned at the top of each account that makes it clear this is a fictional character from a book. That’s a real conversation-stopper, it turns out.
So I have some new followers, but that translated into no sales whatsoever. It didn’t really matter what I tweeted. The only people who starred and RT’d these accounts are people I know from my personal account. All these new followers just don’t care. Twitter analytics showed decent penetration: they were seeing the tweets. They just never interacted with them in any way. And I saw no change in book sales. (It seems that every time someone buys the book they tell me, so I can track effectiveness of marketing quite simply and accurately.)
So I’m declaring the “in character” Twitter experiment a failure. What I’ll do next is simply start tweeting meme images from the book on those accounts. Except I’m going to update them to be a mobile-friendly rectangle shape, instead of the square I used the first time around. I think people might be more interested in the words on the picture if they actually see the words without having to click through. Just a guess.
Another adventure in marketing has to do with Kindle and categories. It’s well known that the categories you choose for your Kindle book have a big impact on sales. If you get into a small category, you’ll have a higher rank, and therefore you’ll get more sales. At least that’s the theory. It’s completely irrelevant in practice for most books, since being ranked anywhere outside the top 100 translates into being completely ignored by Amazon’s recommendation engine. So being #300 in a tiny category or #3000 in a huge one yields the same result: no result.
Nonetheless, I figured I’d play with categories a little to see what I could learn. And it turns out I learned a really important lesson: stay the hell out of the erotica categories! My book isn’t erotica. It has some sex scenes that are more graphic and explicit than your average romance novel. But it isn’t just fuck, fuck, fuckity, fuck. In fact, the sex gets pretty few and far between by the middle of the book. So putting it in erotica seemed kind of silly. On the other hand, conventional wisdom is that the erotica categories (there are two) are smaller than the conventional fiction categories, so I’d get a higher rank.
Setting up your categories is incredibly tricky. You pick two main categories and seven keywords. Amazon munches on these and puts you into different sub-categories using myriad byzantine rules. I started as a contemporary literary romance, basically. I tried switching to BDSM / romantic erotica and saw my rankings plummet. Then I remixed so I’d have both literary and erotica, and my literary rankings disappeared, leaving only the poor erotica rankings. And finally I put it back the way it started and everything went back to normal. After looking at the numbers and reading similar stories on user-to-user forums, I’ve concluded the following:
- The erotica categories “BDSM” and “Romance” are huge, and you will have a low rank in either of them, compared to what you have in non-erotica
- If you choose an erotica category, you are banished from any non-erotica category
- If you choose an erotica category, your overall rank irrespective of category will be one hundred percent worse
That last one was very surprising, but absolutely stood up to my experiments. So the lesson to you authors out there bears repeating: Unless you actually wrote erotica, stay the hell out of the erotica categories!
That’s probably enough for today. I have a whole ‘nother story to tell you about audiobooks. But I’ll leave that as a teaser.
And hey, if you read my book and didn’t leave a review on Amazon, what the fuck? I really need you to do that already. Dammit. In fact, why don’t you do it right now? Here’s a link!